Microsimulation (3).qxd

INTRODUCTION When traffic microsimulation was first introduced commercially in the mid 1990s, it represented a significant advance in modelling. With its focus on vehicle behaviours and interactions on the network, and its ability to test scenarios and immediately see results through a virtual reality display, it was nothing short of a revolution. Since that time traffic microsimulation (hereafter referred to as microsimulation) has advanced in leaps and bounds and has been used for a very wide variety of applications. It is used routinely to examine signalised roundabouts, bus priority, urban traffic control, ramp metering, traffic calming, wide area traffic management, road works design, car park location and control, pedestrian and cyclist interaction, traffic impact, incident management and traffic emissions. In short, it has been used in many circumstances where other modelling systems could not cope. Few would dispute that microsimulation has transformed traffic modelling and revolutionised the way in which the results of modelling are presented to the outside world. To its proponents, the advantages of microsimulation are blindingly obvious, but microsimulation is not without its critics. The most fundamental objection is that any attempt to model individual behaviour is flawed as we do not understand and cannot calibrate the controlling parameters. To the adherents of traditional modelling, microsimulation models are perceived to be expensive, data hungry and almost impossible to validate. This paper originated from the discussion on microsimulation modelling by the Transport Modellers Forum (TMF, formerly TraMPNet), the details of which were reported in the May 2012 edition of Traffic Engineering and Control. It addresses some of the most common criticisms that have been directed at microsimulation, both in terms of concept and application. By reference to a number of practical applications, this paper describes how microsimulation has been applied to tackle and confront issues that, due to their complexity, have been ignored by traditional modelling methods. It concludes by considering the need for better guidance for modellers, focusing on the purpose and application for which the model is required. WHAT IS MICROSIMULATION AND WHY USE IT? Microsimulation differs from traditional highway assignment modelling methods in two distinct ways. Fundamentally, microsimulation models the actions and interactions of individual vehicles, in simulated time steps typically less than 1 second, as they travel through a road network. Traditional models on the other hand assign a matrix of trips to a network calculating average journey times across timeframes of 1 hour or more, using empirical relationships between flow and theoretical capacity. Through its focus on simulating individual vehicles, microsimulation is capable of providing a real time visual display, which represents the second key distinction compared to traditional models. An underlying principle that underpins microsimulation is that the moving components in the system (i.e. the vehicles) perform according to their physical properties and constraints and the behavioural characteristics (aggression and awareness) of their drivers. In microsimulation the movement of individual vehicles is largely governed by three interacting models representing: vehicle following, gap acceptance and lane changing, The results of which vary according to the physical and behavioural attributes in each case. By directly modelling the components of traffic flow instead of their mathematical proxy, microsimulation has provided a unified approach to traffic modelling that has opened up opportunities for modellers to tackle problems where conventional models were found wanting. Accordingly, microsimulation has been applied to a variety of applications ranging from the conventional to the more challenging. For example, microsimulation is ideally suited to model the characteristics of traffic flow on congested motorways, as this relies on an understanding of vehicle behaviour and interactions in order to replicate key flow characteristics such as flow breakdown and shock waves. Furthermore, microsimulation can portray the variable circumstances which lead to congestion in all types and sizes of road network. The significant advances in the presentation and display of results through the provision of a real-time visual display and high resolution 3D graphics afforded by microsimulation has provided a powerful means to communicate the benefits of a scheme with stakeholders and the By Steven Wood